ANALYTICS
LAB
2016
ERIK BRYNJOLFSSON
MIT Initiative on the Digital Economy
http://digital.mit.edu/erik
Information Session
April 13, 2016
1
digital.mit.edu/a-lab
NEW TOOLS BEGET REVOLUTIONS
2
digital.mit.edu/a-lab
1
BIG DATA IS A MEASUREMENT REVOLUTION
•
•
•
•
•
•
•
•
•
•
•
•
3
Clickstream/Page views/Web transactions
Web links/Blog references/Facebook
Google/Bing/Yahoo Searches
Email messages
Mobile phone/GPS/Location data
ERP/CRM/SCM transactions
RFID (Radio Frequency Identification), Bar Code Scanner Data
Real-time machinery diagnostics/engines/equipment
Stock market transactions
Twitter feeds
Wikipedia updates
Etc….
à “Nanodata” and “Nowcasting”
digital.mit.edu/a-lab
BIG DATA IS A MANAGEMENT REVOLUTION
“I think we, as an industry, do a lot of talking... We expect to have
open dialogue. It’s a culture of lunches.
Amazon doesn’t play in that culture. [It has] an incredible discipline
of answering questions by looking at the math, looking at the
numbers, looking at the data. . . .
That’s a pretty big culture clash with the word-and-persuasiondriven lunch culture, the author-oriented culture.”
- Madeline McIntosh, Random House’s President of Sales &
Operations
4
digital.mit.edu/a-lab
2
OVERVIEW
15.572: Analytics Lab: Action Learning Seminar on Analytics, Machine Learning, and the Digital Economy
•
Instructors: Professors Sinan Aral and Erik Brynjolfsson (plus project mentoring team)
•
Schedule: Meets once a week in September and October
•
plus matching workshop in September and final presentations in December (dates TBA)
•
Students from a variety of programs, including MBA, eMBA, SDM, LGO, Sloan Fellow, ORC, MSMS,
EECS, Urban Studies
•
Admission via application, selected based on experience and/or coursework in data science
•
New in 2016, all MBAn students (required coursework)
•
Organization: Student teams of 3-4 design and deliver a project based on the use of analytics,
machine learning, large data sets, or other digital innovations to create or transform a business or
other organization.
•
Many proposals are organizations affiliated with the MIT Initiative on the Digital Economy
•
History
•
A-Lab 2014: 36 students, 10 projects
•
A-Lab 2015: 41 students, 13 projects
•
A-Lab 2016: 60 students, 15-20 projects (estimate)
5
digital.mit.edu/a-lab
EXAMPLES OF PROJECTS
1.
Big Data as a Service (Amazon): Develop demand forecasting of value to Amazon’s retail vendors (2014)
2.
The “Myth of the Crystal Ball”: Understanding Forecasting Errors at Amazon (Amazon): Quantify the
impact of supply chain forecasting errors to better prioritize forecast improvements in the future (2015)
3.
Understanding Supply and Demand in the Boston Public Schools (Boston Public Schools): Use the
BPS student dataset to generate hypotheses about what drives demand for schools in the Boston area,
helping BPS to "right-size" school districts (2015)
4.
Populating "Popular Now": Rebooting our News Story Recommendation Algorithm (Christian
Science Monitor): Develop a news recommendation algorithm to drive page views and user engagement
on the Christian Science Monitor site. Try to beat the existing "Popular Now" algorithm. (2015)
5.
Understanding Successful eBay Sale Prices (eBay): Find the factors that best predict successful prices
for new and used eBay items in different categories and under a variety of sales conditions (2015)
6.
Predicting Hospital Readmission (Dell): Find the factors that best predict 30-day hospital readmission
(2015)
6
digital.mit.edu/a-lab
3
EXAMPLES OF PROJECTS (CONTINUED)
7.
Finding the Next Watson Use Case (IBM Watson): Case chosen: compliance by financial institutions
with federal regulations (2014)
8.
Identifying Fraud for an Online Gift Card Platform (Raise Marketplace): Develop an algorithm to help
Raise classify transactions as fraudulent or legitimate (2015)
9.
Predictive Maintenance in the Elevator and Escalator Industry (Schindler Elevator): Help Schindler
use predictive analytics to revise its maintenance strategy and better perform preventative intervention
(2015)
10.
Using Geospatial Data to Develop a New Kind of Football Analytics (Telemetry Sports): Use a new
source of geospatial NFL data to classify plays, evaluate players, and design football strategy (2015)
11.
Multi-channel Consumer Profiling for eCommerce (WOOX): Provide more segmentation and profiles of
potential customers for WOOX’s high quality headphones (2014)
12.
Predicting New Product Adoption for American Apparel (Zensar): Sponsor challenge: “We may have
people with experience, wisdom, and opinions, predicting sales of a new line of jeans. Can we do better
with analytics?” (2014)
7
digital.mit.edu/a-lab
A DEEPER DIVE – AMAZON (2015)
The “Myth of the Crystal Ball”: Understanding Forecasting Errors at Amazon
Challenge: Help Amazon quantify the impact of supply chain forecasting errors to
better prioritize forecast improvements in the future.
Data: 75 million rows containing daily demand and forecast data for 206 thousand
products over two weeks.
Analysis: Defined different kinds of costs associated with forecasting errors and
their magnitudes. Used statistical methods in R running on a cloud computing
system to quantify lost profit due to forecast error.
Recommendation: Incorporate indirect costs into the evaluation of forecasting
errors. Look for variation across product categories.
8
digital.mit.edu/a-lab
4
SLOAN COURSES WITH ANALYTICS CONTENT
15.034 Metrics for Managers: Big Data and Better Answers (Doyle)
15.564 IT Essentials II: Advanced Technologies
Economy (Madnick)
for Digital Business in the Knowledge
15.060 Data, Models, and Decisions (Bertimas, et al)
15.565J Digital Evolution:
15.062J Data Mining: Finding the Data and Models that Create Value (Welsch)
15.567 The Economics
Managing Web 3.0 (Madnick)
of Information:
Strategy, Structure, and Pricing (Brynjolfsson)
15.071 The Analytics Edge (Bertsimas)
15.569 Leadership Lab: Leading Sustainable
15.074J Predictive Analytics and Statistical Modeling (Welsch)
Systems (Senge, Orlikowski)
15.570 Digital Marketing and Social Media Analytics (Aral)
15.075 Statistical Thinking and Data Analysis (Rudin)
15.571 Enterprise Transformations
15.096 Prediction: Machine Learning and Statistics (Rudin)
15.575 Economics
(Brynjolfsson)
in the Digital Economy
of Information and Technology
(Ross)
in Markets and Organiz ations
15.320 Strategic Organiz ational Design (Malone)
15.576 Research Seminar
in IT and Organiz ations: Social Perspectives (Orlikowski)
15.339 Distributed Leadership Workshop (Ancona, Malone, Orlikowski)
15.376J Media Ventures (Pentland, Bonsen)
15.578 Global Information Systems: Strategic, Technical, and Organiz ational
Perspectives (Madnick)
15.377J Linked Data Ventures (Berners-Lee, Kagal, Rae, Sturdevant)
15.579-15.580 Seminar in Information Technology
15.561 Information Technology
(Madnick, Malone, Orlikowski)
Essentials (Malone)
9
digital.mit.edu/a-lab
HOW TO APPLY: SELECTIVE ADMISSION
• Open to Sloan MBAs, eMBAs, and other MIT graduate
students
• Application available 12:00pm, May 2 through 12:00pm
May 9 on digital.mit.edu/a-lab
• Notifications of admission decision will be sent in midMay
• No bidding for 15.572 is necessary
10
digital.mit.edu/a-lab
5
CLOSING THOUGHT
“Technological progress is going to leave
behind some people, perhaps even a lot
of people, as it races ahead.
But there’s never been a better time to be
a worker with special skills or the right
education, because these people can use
technology to create and capture value”
The Second Machine Age, p 11.
11
digital.mit.edu/a-lab
THE INITIATIVE ON THE DIGITAL ECONOMY
http://mitsloan.mit.edu/ide
12
digital.mit.edu/a-lab
6
QUESTIONS?
For questions about the course, please contact
Susan Young susany@mit.edu
More information is also available at:
digital.mit.edu/a-lab (course site) and
http://stellar.mit.edu/S/course/15/fa15/15.572/ (MIT Stellar site)
13
digital.mit.edu/a-lab
7